PP53B-01 INVITED
Implications of contemporaneous observations of glacier changes on the sensitivity of ice sheets to climate change.
Modern observations of ice sheet evolution, gathered mostly from satellites, have shed new lights on the sensitivity of glaciers and ice sheets to climate change. Widespread glacier acceleration observed along the southern coast of Greenland doubled the ice sheet mass deficit within a decade. The acceleration resulted from the thinning of the frontal regions by warm temperatures, which reduced buttressing and allowed faster rates of ice flow. Glaciers grounded well below sea level respond more dramatically to thinning as buoancy forces provide a positive feedback to the glacier retreat into deeper waters. Surface melt water and warmer ocean also play a direct role on the glacier evolution, but this is not fully understood at present. In the Antarctic Peninsula, there is widespread glacier acceleration on the west coast similar to that observed in Greenland. On the east coast, abrupt changes caused by ice-shelf collapses generated ten-fold increases in glacier flow which dwarf changes observed in Greenland. These changes in ice dynamics are not only significant in terms of mass balance, they dominate and make this region a contributor to sea level rise comparable to Alaska and Patagonia. Farther south, in the Bellingshausen and Amunden sea sectors, where climate warming is more subdued, there is widespread ice sheet imbalance concentrated along the narrow channels occupied by glaciers, hence also caused by ice dynamics. These glaciers are grounded well below sea level and prone to rapid retreat. Glacier thinning of many meters per year is common.Glacier and ice- shelf thinning in these regions are controlled by the thermal forcing from the ocean (ablation rates of ice shelves are orders of magnitude larger than surface mass balance) and the depth of the glaciers below sea level.Unfortunately, we lack of a lot of basic information on both. Yet, we know enough to recognize that numerical models of ice sheet flow that do not include glacier dynamics and ice shelf/ice stream interactions cannot explain any of the contemporaneous observations of glacier change. Over time scales of centuries, which are of interest to predict what ice sheets will do next, more sophisticated models that include glacier dynamics, thermal forcing from the ocean, and a detailed mapping of depths below sea level are essential. The International Polar Year will help make timely, major advances in this direction. This work was performed at Caltech's Jet Propulsion Laboratory under a contract with the National Aeronautics and Space Administration.
PP53B-02 INVITED
Rapid increase in global ice volume at the inception of the Last Glacial Maximum
The timing and magnitude of global ice volume changes during the transition between Marine Isotope Stage (MIS) 3 to 2 (ie. LGM) is poorly constrained as relevant sea-level indicators are presently located underwater. Drill cores from buried coral terraces, dated by U-series nuclides, have provided some data, however, additional data are required to reconstruct a reliable global ice volume history for this period. Japan Sea is separated from the open ocean by shallow sills (water depth ranges from 15-135m) and its hydrology is affected by global sea-level changes. We obtained twenty AMS (Accelerator Mass Spectrometry) radiocarbon dates, 2 tephra layer chronologies, and planktonic foraminifera d18O stratigraphy for the last 50,000 years from a sediment core recovered from Oki ridge. Radiocarbon ages of the tephra layers indicate that the Japan Sea reservoir ages for the Holocene and the last glacial period were similar. Organic carbon and nitrogen isotopes were measured throughout the core with a temporal resolution as good as ca. 200 years. Carbon and nitrogen isotope ratios during the Holocene remained remarkably constant and the shift from the LGM towards the Holocene occurred at ca. 19 cal ka. Carbon isotopes during the LGM were also stable and the transition from 2 per mil heavier carbon to LGM values was evident at about 28 cal ka. A similar shift is present in the nitrogen isotopes and points to major environmental changes at this time. We have also measured oxygen isotopes in planktonic forams which show a freshening in the salinity of the Japan Sea, compared to present, during periods of low sea-level. The timing of the changes in d18O also correspond to the shifts in the C and N isotopes. Therefore, it is reasonable to assume that the cause of the observed environmental changes in the Japan Sea can be attributed directly to the rapid fall in sea-level from MIS-3 to the LGM. Our records, combined with global sea-level data, suggest that the LGM started at about 30 ka with almost 50m fall in sea-level in 2000 years. The timing and magnitude of the LGM inception is consistent with coral records from Huon Peninsula and Barbados.
PP53B-03 INVITED
Searching for Eustasy in Deglacial Sea Level Histories
Relative sea-level (RSL) histories over the last glacial cycle may be decomposed into a eustatic (i.e., globally uniform) term and a far more complex signal associated with glacial isostatic adjustment (GIA). The spatial and temporal complexity of the latter arises from deformational and gravitational perturbations driven by the surface mass (ice plus ocean) redistribution. Any attempt to accurately estimate excess ice volume since the LGM reduces to an effort to isolate the eustatic signal, either by choosing to analyze RSL records at a geographic site where GIA effects are thought to be small or where these effects can be removed using numerical models of post-glacial sea-level change. In this talk we present a large suite of global, deglacial sea-level predictions generated by varying parameters defining both the Late Pleistocene ice history and viscoelastic Earth model. Our calculations are based on a new generation of post-glacial sea-level algorithms that take accurate account of shoreline migration processes and Earth rotation, as well as 3-D Earth structure. We use the model predictions to generate global maps of the mean departure of the sea-level histories from the eustatic signal and a standard deviation that reflects reasonable uncertainties in the input (ice, Earth) parameters. Our results demonstrate that there are relatively few locations where: (I) the total predicted RSL change closely tracks the eustatic value; and (II) the GIA component of the signal is insensitive to reasonable variations in the input parameters (and so can be accurately removed from the observations). The maps will be a useful tool for field scientists in choosing suitable sites for future data acquisition.
PP53B-04
Millennial Climate Oscillations and Global sea Level Changes. Evidence of Rapid Flooding Events in the Western Mediterranean Continental Margin
A continuous sediment record recovered in Borehole PRGL1 drilled in the upper slope from the Western Mediterranean continental margin allowed for the first time to directly calibrate millennial global sea level records to Greenland ice core chronologies. This record shows a series of condensed layers formed at times of rapid flooding of the continental shelf, when coastline moved landward tens of kms in a few hundreds of years. As the coastline moved landward sedimentation rate in the upper slope abruptly decreased with a simultaneous change from fine-grained, deltaic sediments to coarse-grained biogenic deposits. During the last deglaciation a prominent condensed layer was formed, which was dated between 14.5 kyr and the present. Other condensed layers formed during the major deglaciations of the last 400 kyr. Between 70 and 20 kyr five meltwater events were recognized that were calibrated to Greenland chronologies through the comparison of oxygen isotope records from the Mediterranean and Greenland ice cores. Rapid changes in Mediterranean sea surface temperatures simultaneous with Dansgaard-Oeschger events in Greenland allowed us to establish accurate phase relationships between millennial global sea level changes, variations in CO2 and abrupt climate oscillations in both Hemispheres.
PP53B-05
A 3-million-year reconstruction of climate, ice volume and sea level; identifying mechanisms behind the inception of Northern-Hemisphere glaciation and the mid-Pleistocene transition
A coupled ice-sheet ocean-temperature model was used to extract three-million-year records of surface air temperature, ice volume, and sea level from deep-sea oxygen isotopes. The results show that a 5 °C cooling between 3 and 2.5 million years ago (Ma) initiated Northern Hemispheric glaciation around 2.7 Ma with low amplitude obliquity-scale (41-kyr) glacials - dominated by the sluggish Eurasian ice sheets - prevailing. Prior to about 1 Ma, ice volume lagged behind air temperature by 5 kyr in quasi-linear 41-kyr cycles. Ongoing climate cooling enabled the Cordilleran and Laurentide ice sheets to survive insolation maximums, causing them to eventually merge and to expand rapidly during glacial conditions. Once the huge North-American ice sheet reached a certain extent and/or thickness, the subsequent insolation maximum-related warming then initiated deglaciation, presumably through basal sliding related instabilities. The 100-kyr glacials arose gradually between 1.4 and 0.6 Ma, with the different response time and geometrical constraints of the Eurasian and North-American ice sheets being key factors.
PP53B-06
Global sea-level changes inferred from U.S. margin and deep-sea isotopic records
U.S. mid-Atlantic margin records provide constraints on interpretations of global sea-level and deep-sea isotopic records. Based on our scaling of oxygen isotopic records, we predict preservation of strata only during peak interglacials (e.g., 5e, 11) back through the Pliocene. Previous studies in Virginia and the Carolinas suggest much higher sea level than today during the early Pliocene (35±15m); the absence of Pliocene strata in New Jersey and Delaware suggest either sea-level lower than modern or uplift of this region relative to Virginia and the Carolinas. Backstripping of Virginia records suggest a maximum eustatic highstand less than 20 m during the early Pliocene consistent with our scaling of deep-sea isotope records. The margin was largely in hiatus from ca. 2 Ma to stage 5e (125 ka) suggesting sea-levels below present; two likely exceptions were ca. 400 ka and ca. 1 Ma highstands that probably correlate with stages 11 and 31 mega-interglacials. Stages 5e and the Holocene are recorded onshore in New Jersey, whereas other late Pleistocene stages are only found offshore. The late Pleistocene sea-level record from the U.S. middle Atlantic margin is similar to records from New Guinea, Barbados, Araki, and the Red Sea for stages 1, 2, 4, 5e, and 6; large differences exist among records for MIC 3, with smaller differences in MIC5a and-5c. Holocene sea-level records from the Mid-Atlantic region show surprising uniformity considering different proximities to the peripheral bulge, with a relative rise throughout the region of ~1.7-1.9 mm/y since ~5000 yBP. Correcting for geoidal subsidence, the U.S. east coast records suggest a global sea-level (eustatic) rise of ~0.8 mm/y since 5000 yBP. Comparison with other records provides a best estimate of pre-anthropogenic global sea-level rise of 0.8±0.3 mm/y from 5000 until ~1750 AD, indicating that anthropogenic influences are far larger than the natural rise.